Last Updated: 27 July 2026
Multiverse Computing, a Spain-based startup that compresses large AI models so they can run on everyday devices and energy-efficient data centers, has raised $570 million (€500 million) in a Series C round. Co-founded by CEO Enrique Lizaso and Chief Scientific Officer Dr. Román Orús, the company applies techniques borrowed from quantum physics to shrink AI models by up to 95% while preserving their accuracy.
Running powerful AI today is expensive. Training and operating large language models (LLMs) requires enormous computing power and energy, and the bulk of that processing takes place inside a handful of hyperscale data centers, most of them in the United States. For governments, manufacturers, and enterprises that want to run AI locally, whether on their own servers, on a factory floor, or inside a smartphone, the cost and infrastructure demands have been a real barrier. The edge AI market, which covers AI processing done on or near a device rather than in a remote cloud, was valued at roughly $25 billion in 2025 and is projected to grow at more than 20% annually through the end of the decade, driven by demand from manufacturing, automotive, telecom, and defense.
The round values Multiverse at $1.7 billion pre-money, a fivefold step up from its Series B. Forgepoint Capital International, BNPP SIVF, and Bullhound Capital co-led the investment. Other participants include Santander Alternative Investments, Tikehau Capital, HP Inc., Orange Ventures, Scania Invest, NAventures, Qatar Development Bank, Zouk Capital, SETT (a Spanish state-owned enterprise), the EIC Fund, the Basque Government's Hazten Scale-Up Fund, and Kutxa Fundazioa, among others. JP Morgan and Santander CIB advised on the round, which is expected to bring total funding to $800 million and may remain open to additional strategic investors.
This investment arrives as both industry and governments are rethinking where and how AI infrastructure should be built. In the United States, the major cloud providers have committed hundreds of billions of dollars to ever-larger data centers. Europe is moving differently. The European Commission's InvestAI initiative, formalized in early 2026, targets €200 billion in total AI investment through 2030, including plans for up to five AI gigafactories, which are large-scale computing facilities designed for training and running AI models. Still, Europe lags well behind the U.S. and China in data center capacity, and many European organizations continue to depend on American cloud providers for their most demanding AI workloads.
That gap has made "sovereign AI" a pressing policy concern. The term refers to a country or region's ability to run AI systems on its own infrastructure, under its own rules, without relying on foreign providers. In sectors like defense, energy, healthcare, and finance, where data sensitivity and regulatory compliance are paramount, this is a practical necessity, not just a political talking point. Multiverse's argument is straightforward: by making AI models dramatically smaller and more efficient, organizations and governments can deploy them locally, using less hardware and less energy.
"Beyond delivering substantial cost savings for its clients, Multiverse's solutions achieve a structural reduction in the energy intensity of LLM workloads through advanced model‑compression techniques. This translates into lower GPU utilization, reduced energy consumption, and decreased infrastructure requirements, all while preserving comparable performance and accuracy levels. BNP Paribas SIVF is delighted to back Multiverse, whose transformative technology promises a significant step forward for the sustainable use of resources in AI model inference."
Yann Lagalaye, Managing Partner at BNPP SIVF
The Series C will fund several priorities. Multiverse plans to expand its library of compressed AI models, continue developing its proprietary compression algorithms, and invest in sovereign AI gigafactory infrastructure along with the software needed to support those deployments. The company also intends to build a stronger presence in East Asia, Southeast Asia, the Middle East, Canada, and the United States.
Commercially, the numbers have been moving fast. Since closing its Series B in June 2025, Multiverse says annualized revenue has grown by more than ten times, with 96x year-over-year sales growth in the first quarter of 2026. Its models are already deployed across millions of devices, including drones, cameras, satellites, vehicles, and telecom infrastructure, and are set to be embedded into AI-equipped personal computers.
"The AI industry has built an extraordinary capability on top of an infrastructure model that wasn't designed to sustain it. The winners in the next phase of this multi-trillion-dollar market will be the ones that make every data center, every GPU, and every enterprise deployment dramatically more efficient. Multiverse sits at the intersection of the infrastructure and the application layers and has evolved from being the leading downstream LLM compression technology to becoming a complete AI foundry and Operating System. It is the only company we've seen that has both the technical foundation and the commercial traction to be that critical platform."
Damien Henault, Managing Director and Partner at Forgepoint Capital International
Multiverse Computing was founded in 2019 and is headquartered in Donostia (San Sebastián), Spain. It started as a quantum software company before turning its research toward making AI models more efficient. At the center of its platform is CompactifAI, a compression technology built on tensor networks, a mathematical framework originally developed for quantum physics. The result: large language models can be reduced in size by 80% to 95% while keeping accuracy largely intact. Think of it this way. A large AI model that normally needs a powerful server to run gets compressed until it can operate on a smartphone, a drone, or a small on-site server.
CompactifAI supports the full range of deployment environments. Compressed models can run in the cloud, on a company's own servers, or directly on devices. A routing system called the CompactifAI Router decides in real time whether a given task should be handled locally or sent to the cloud. For data centers, Multiverse offers what it describes as a software layer that combines model compression, GPU orchestration, and sovereign-grade controls into a single platform. Customers and partners include Allianz, Bank of Canada, Bosch, Iberdrola, Indra, PwC, and Telefónica, spanning industries from finance and energy to defense and health sciences. The company serves more than 100 customers and has offices across Europe, the United States, and Canada.
"The AI industry has accepted a false constraint for years — that powerful models require expensive infrastructure. That constraint is gone. We have proven that AI can run at full performance on a smartphone, inside a sovereign data center, on a factory floor with no cloud connection. This round is the moment we scale that proof across every industry that needs it."
Enrique Lizaso, Co-founder and CEO of Multiverse Computing
The $570 million round is expected to bring Multiverse Computing's total funding to $800 million, inclusive of prior rounds. Forgepoint Capital International, a venture firm focused on cybersecurity, artificial intelligence, and infrastructure software, co-led alongside BNPP SIVF and Bullhound Capital. The investor roster spans sovereign-linked capital from Qatar Development Bank and Spain's SETT, participation from HP Inc., and corporate venture arms including Orange Ventures, Scania Invest, and NAventures (National Bank of Canada). The European Innovation Council Fund and the Basque Government's Hazten Scale-Up Fund also took part.
Multiverse's previous Series B closed in June 2025. Specific details of that round were not disclosed, but the Series C valuation of $1.7 billion represents a fivefold increase. JP Morgan and Santander CIB handled the advisory, and the round may remain open to select additional strategic investors.
"For years, Europe has worried it cannot compete at the frontier of AI. Multiverse is the answer — a team out of San Sebastián whose technology lets anyone run powerful models on their own terms, on their own soil, without depending on anyone else. Backing Román, Enrique, and this team was one of the most important decisions we have made."
Per Roman, Co-founder and Managing Partner at Bullhound Capital
